Abstract
Protecting the rights of inventor is not the only purpose of patents, and it is becoming more and more important under the keen competitive business environment nowadays. In the patent systems of many countries, patent holders are required to pay a maintenance fee after the initial application to retain patent protection on any invention until the expiration of the protection period. However, not all patents are worth maintaining by patent holders. Thus, organizations need to identify “important patents” for maintenance and abandon “unimportant patents.” In this study, we attempt to examine factors influencing patent maintenance via a Delphi study and propose a data mining approach to support patent maintenance decision. Such a data-mining-based patent maintenance decision support system can help organizations improve the effectiveness of patent maintenance decisions and, at the same time, decrease the cost of maintenance decision. Our empirical evaluation results suggest that our proposed model outperforms the benchmark model (i.e., involving the variables suggested by the literature only).